To address errors stemming from the dynamics model and torque control accuracy, as well as to overcome limitations in whole-body kinematics, we introduce a novel hierarchical hybrid whole-body control framework tailored for wheel-legged robots. This framework comprises two pivotal modules: reduced-order motion planning and hybrid whole-body control. In motion planning, we leverage a simplified model characterized by broad linear range capabilities to maintain sagittal balance. Within the hybrid control module, we integrate a multi-rigid body model encompassing full dynamics and kinematics and devise two optimization solvers to compute optimal torques and joint positions. Finally, the comparative simulations were conducted to validate the efficacy and robustness of the proposed hybrid controller in terms of anti-interference, tracking accuracy, and adaptability to uneven terrain.